Presentation Information
[10p-E215-11]Automated Analysis of Domain Distribution in CVD Graphene by Photoelectron Momentum Microscopy
Shinichiro Tanaka1, Shigemasa Suga1, Yuki Somei2, Yusuke Sato2, Fumihiko Matsui2, Itsuki Imahori3, Yoshihide Aoyagi3, 〇SHINYA OHNO3 (1.SANKEN Osaka Univ., 2.UVSOR, IMS, 3.Yokohama Nat'l Univ.)
Keywords:
Graphene,Photoelectron Momentum Microscope,Automated Analysis
Recent advances in electron spectroscopy have led to a rapid increase in data volume, making conventional manual analysis increasingly impractical. In this study, we developed an automated method to extract crystallographic orientation distributions of CVD graphene by applying a PyTorch-based optimization algorithm to angle-resolved datasets acquired with the photoelectron momentum microscope at UVSOR. By modeling the hexagonal momentum-space pattern and optimizing parameters via gradient descent using cosine similarity as the loss function, we successfully constructed high-precision orientation maps and histograms. The resulting distributions provide important insights into nucleation and domain formation mechanisms during polycrystalline graphene growth.